{"id":29571829,"url":"https://github.com/epic-kitchens/epic-kitchens-55-annotations","last_synced_at":"2025-07-19T04:36:17.138Z","repository":{"id":41387280,"uuid":"126323601","full_name":"epic-kitchens/epic-kitchens-55-annotations","owner":"epic-kitchens","description":"🍴 Annotations for the EPIC KITCHENS-55 Dataset.","archived":false,"fork":false,"pushed_at":"2021-03-17T13:01:03.000Z","size":30833,"stargazers_count":151,"open_issues_count":4,"forks_count":26,"subscribers_count":13,"default_branch":"master","last_synced_at":"2025-05-21T10:40:04.922Z","etag":null,"topics":["action-recognition","dataset","deep-learning","egocentric","egocentric-action-recognition","first-person-video","object-detection","video"],"latest_commit_sha":null,"homepage":"https://epic-kitchens.github.io/2020","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/epic-kitchens.png","metadata":{"files":{"readme":"README.html","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2018-03-22T11:14:03.000Z","updated_at":"2025-04-25T06:55:13.000Z","dependencies_parsed_at":"2022-09-19T10:30:54.699Z","dependency_job_id":null,"html_url":"https://github.com/epic-kitchens/epic-kitchens-55-annotations","commit_stats":null,"previous_names":[],"tags_count":14,"template":false,"template_full_name":null,"purl":"pkg:github/epic-kitchens/epic-kitchens-55-annotations","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/epic-kitchens%2Fepic-kitchens-55-annotations","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/epic-kitchens%2Fepic-kitchens-55-annotations/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/epic-kitchens%2Fepic-kitchens-55-annotations/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/epic-kitchens%2Fepic-kitchens-55-annotations/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/epic-kitchens","download_url":"https://codeload.github.com/epic-kitchens/epic-kitchens-55-annotations/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/epic-kitchens%2Fepic-kitchens-55-annotations/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":265889155,"owners_count":23844539,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["action-recognition","dataset","deep-learning","egocentric","egocentric-action-recognition","first-person-video","object-detection","video"],"created_at":"2025-07-19T04:36:16.479Z","updated_at":"2025-07-19T04:36:17.115Z","avatar_url":"https://github.com/epic-kitchens.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003ch1 id=\"epic-kitchens-55-dataset\"\u003e\u003ca href=\"https://epic-kitchens.github.io/2018\"\u003eEPIC KITCHENS-55 Dataset\u003c/a\u003e\u003c/h1\u003e\n\u003c!-- start badges --\u003e\n\u003c!-- end badges --\u003e\n\u003cblockquote\u003e\n\u003cp\u003e\u003ca href=\"https://epic-kitchens.github.io/\"\u003eEPIC-KITCHENS-55\u003c/a\u003e is the largest dataset in first-person (egocentric) vision; 55 hours of multi-faceted, non-scripted recordings in native environments - i.e. the wearers’ homes, capturing all daily activities in the kitchen over multiple days. Annotations are collected using a novel `live’ audio commentary approach.\u003c/p\u003e\n\u003c/blockquote\u003e\n\u003ch2 id=\"authors\"\u003eAuthors\u003c/h2\u003e\n\u003cp\u003eDima Damen (1) Hazel Doughty (1) Giovanni Maria Farinella (3) Sanja Fidler (2) Antonino Furnari (3) Evangelos Kazakos (1) Davide Moltisanti (1) Jonathan Munro (1) Toby Perrett (1) Will Price (1) Michael Wray (1)\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e(1 University of Bristol)\u003c/li\u003e\n\u003cli\u003e(2 University of Toronto)\u003c/li\u003e\n\u003cli\u003e(3 University of Catania)\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003eContact:\u003c/strong\u003e \u003ca href=\"mailto:uob-epic-kitchens@bristol.ac.uk\"\u003euob-epic-kitchens@bristol.ac.uk\u003c/a\u003e\u003c/p\u003e\n\u003ch2 id=\"citing\"\u003eCiting\u003c/h2\u003e\n\u003cp\u003eWhen using the dataset, kindly reference:\u003c/p\u003e\n\u003cpre\u003e\u003ccode\u003e@INPROCEEDINGS{Damen2018EPICKITCHENS,\n   title={Scaling Egocentric Vision: The EPIC-KITCHENS Dataset},\n   author={Damen, Dima and Doughty, Hazel and Farinella, Giovanni Maria  and Fidler, Sanja and \n           Furnari, Antonino and Kazakos, Evangelos and Moltisanti, Davide and Munro, Jonathan \n           and Perrett, Toby and Price, Will and Wray, Michael},\n   booktitle={European Conference on Computer Vision (ECCV)},\n   year={2018}\n} \u003c/code\u003e\u003c/pre\u003e\n\u003cp\u003e(Check publication \u003ca href=\"https://epic-kitchens.github.io\"\u003ehere\u003c/a\u003e)\u003c/p\u003e\n\u003ch2 id=\"dataset-details\"\u003eDataset Details\u003c/h2\u003e\n\u003ch3 id=\"ground-truth\"\u003eGround Truth\u003c/h3\u003e\n\u003cp\u003eWe provide ground truth for action segments and object bounding boxes.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\u003cstrong\u003eObjects:\u003c/strong\u003e Full bounding boxes of narrated objects for every annotated frame.\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003eActions:\u003c/strong\u003e Split into narrations and action labels:\n\u003cul\u003e\n\u003cli\u003eNarrations containing the narrated sentence with the timestamp.\u003c/li\u003e\n\u003cli\u003eAction labels containing the verb and noun labels along with the start and end times of the segment.\u003c/li\u003e\n\u003c/ul\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"dataset-splits\"\u003eDataset Splits\u003c/h3\u003e\n\u003cp\u003eThe dataset is comprised of three splits with the corresponding ground truth:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eTraining set - Full ground truth.\u003c/li\u003e\n\u003cli\u003eSeen Kitchens (S1) Test set - Start/end times only.\u003c/li\u003e\n\u003cli\u003eUnseen Kitchens (S2) Test set - Start/end times only.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eInitially we are only releasing the full ground truth for the training set in order to run action and object challenges.\u003c/p\u003e\n\u003ch3 id=\"important-files\"\u003eImportant Files\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ccode\u003eREADME.md (this file)\u003c/code\u003e\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003eREADME.html\u003c/code\u003e\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003eREADME.pdf\u003c/code\u003e\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"#license\"\u003e\u003ccode\u003elicense.txt\u003c/code\u003e\u003c/a\u003e\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_train_action_labels.csv\"\u003e\u003ccode\u003eEPIC_train_action_labels.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_train_action_labelscsv\"\u003eInfo\u003c/a\u003e) (\u003ca href=\"EPIC_train_action_labels.pkl\"\u003ePickle\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_train_object_labels.csv\"\u003e\u003ccode\u003eEPIC_train_object_labels.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_train_object_labelscsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_test_s1_timestamps.csv\"\u003e\u003ccode\u003eEPIC_test_s1_timestamps.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_test_s1_timestampscsv\"\u003eInfo\u003c/a\u003e) (\u003ca href=\"EPIC_test_s1_timestamps.pkl\"\u003ePickle\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_test_s2_timestamps.csv\"\u003e\u003ccode\u003eEPIC_test_s2_timestamps.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_test_s2_timestampscsv\"\u003eInfo\u003c/a\u003e) (\u003ca href=\"EPIC_test_s2_timestamps.pkl\"\u003ePickle\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_train_object_action_correspondence.csv\"\u003e\u003ccode\u003eEPIC_train_object_action_correspondence.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_train_object_action_correspondencecsv\"\u003eInfo\u003c/a\u003e) (\u003ca href=\"EPIC_train_object_action_correspondence.pkl\"\u003ePickle\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_test_s1_object_action_correspondence.csv\"\u003e\u003ccode\u003eEPIC_test_s1_object_action_correspondence.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_test_s1_object_action_correspondencecsv\"\u003eInfo\u003c/a\u003e) (\u003ca href=\"EPIC_test_s1_object_action_correspondence.pkl\"\u003ePickle\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_test_s2_object_action_correspondence.csv\"\u003e\u003ccode\u003eEPIC_test_s2_object_action_correspondence.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_test_s2_object_action_correspondencecsv\"\u003eInfo\u003c/a\u003e) (\u003ca href=\"EPIC_test_s2_object_action_correspondence.pkl\"\u003ePickle\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_test_s1_object_video_list.csv\"\u003e\u003ccode\u003eEPIC_test_s1_object_video_list.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_test_s1_object_video_listcsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_test_s2_object_video_list.csv\"\u003e\u003ccode\u003eEPIC_test_s2_object_video_list.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_test_s2_object_video_listcsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_noun_classes.csv\"\u003e\u003ccode\u003eEPIC_noun_classes.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_noun_classescsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_verb_classes.csv\"\u003e\u003ccode\u003eEPIC_verb_classes.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_verb_classescsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3 id=\"additional-files\"\u003eAdditional Files\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"EPIC_train_invalid_labels.csv\"\u003e\u003ccode\u003eEPIC_train_invalid_labels.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_train_invalid_labelscsv\"\u003eInfo\u003c/a\u003e) (\u003ca href=\"EPIC_train_invalid_labels.pkl\"\u003ePickle\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_train_action_narrations.csv\"\u003e\u003ccode\u003eEPIC_train_action_narrations.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_train_action_narrationscsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_descriptions.csv\"\u003e\u003ccode\u003eEPIC_descriptions.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_descriptionscsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_many_shot_verbs.csv\"\u003e\u003ccode\u003eEPIC_many_shot_verbs.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_many_shot_verbscsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_many_shot_nouns.csv\"\u003e\u003ccode\u003eEPIC_many_shot_nouns.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_many_shot_nounscsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_many_shot_actions.csv\"\u003e\u003ccode\u003eEPIC_many_shot_actions.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_many_shot_actionscsv\"\u003eInfo\u003c/a\u003e)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"EPIC_video_info.csv\"\u003e\u003ccode\u003eEPIC_video_info.csv\u003c/code\u003e\u003c/a\u003e (\u003ca href=\"#epic_video_infocsv\"\u003einfo\u003c/a\u003e)\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eWe direct the reader to \u003ca href=\"https://data.bris.ac.uk/data/dataset/3h91syskeag572hl6tvuovwv4d\"\u003eRDSF\u003c/a\u003e for the videos and rgb/flow frames.\u003c/p\u003e\n\u003cp\u003eWe provide html and pdf alternatives to this README which are auto-generated.\u003c/p\u003e\n\u003ch2 id=\"files-structure\"\u003eFiles Structure\u003c/h2\u003e\n\u003ch3 id=\"epic_train_action_labels.csv\"\u003eEPIC_train_action_labels.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 14 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 15%\" /\u003e\n\u003ccol style=\"width: 8%\" /\u003e\n\u003ccol style=\"width: 65%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003euid\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e6374\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eUnique ID of the segment.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP03_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo the segment is in.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enarration\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eclose fridge\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnglish description of the action provided by the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:23:43.847\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:23:47.212\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e85430\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e85643\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP03\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everb\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eclose\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eParsed verb from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003efridge\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFirst parsed noun from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everb_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e3\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the parsed verb’s class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e10\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the parsed noun’s class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eall_nouns\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003elist of string (1 or more)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e['fridge']\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eList of all parsed nouns from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eall_noun_classes\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003elist of int (1 or more)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e[10]\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eList of numeric IDs corresponding to all of the parsed nouns’ classes from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePlease note we have included a python pickle file for ease of use. This includes a pandas dataframe with the same layout as above. This pickle file was created with pickle protocol 2 on pandas version 0.22.0.\u003c/p\u003e\n\u003ch3 id=\"epic_train_invalid_labels.csv\"\u003eEPIC_train_invalid_labels.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 14 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 15%\" /\u003e\n\u003ccol style=\"width: 8%\" /\u003e\n\u003ccol style=\"width: 65%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003euid\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e6374\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eUnique ID of the segment.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP03_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo the segment is in.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enarration\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eclose fridge\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnglish description of the action provided by the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:23:43.847\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:23:47.212\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e85430\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e85643\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP03\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everb\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eclose\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eParsed verb from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003efridge\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFirst parsed noun from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everb_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e3\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the parsed verb’s class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e10\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the parsed noun’s class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eall_nouns\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003elist of string (1 or more)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e['fridge']\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eList of all parsed nouns from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eall_noun_classes\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003elist of int (1 or more)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e[10]\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eList of numeric IDs corresponding to all of the parsed nouns’ classes from the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePlease note we have included a python pickle file for ease of use. This includes a pandas dataframe with the same layout as above. This pickle file was created with pickle protocol 2 on pandas version 0.22.0.\u003c/p\u003e\n\u003ch3 id=\"epic_train_action_narrations.csv\"\u003eEPIC_train_action_narrations.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 5 columns:\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNote: The start/end timestamp refers to the start/end time of the narration, not the action itself.\u003c/em\u003e\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 17%\" /\u003e\n\u003ccol style=\"width: 6%\" /\u003e\n\u003ccol style=\"width: 15%\" /\u003e\n\u003ccol style=\"width: 62%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP03\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP03_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo the segment is in.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:23:43.847\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:23:47.212\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the narration.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enarration\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eclose fridge\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnglish description of the action provided by the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_train_object_labels.csv\"\u003eEPIC_train_object_labels.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 6 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 11%\" /\u003e\n\u003ccol style=\"width: 19%\" /\u003e\n\u003ccol style=\"width: 17%\" /\u003e\n\u003ccol style=\"width: 51%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e20\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eInteger value representing the class in noun-classes.csv.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003ebag\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eOriginal string name for the object.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo the object was annotated in.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eframe\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e056581\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame number of the annotated object.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003ebounding_boxes\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003elist of 4-tuple (0 or more)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e\u0026quot;[(76, 1260, 462, 186)]\u0026quot;\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eAnnotated boxes with format \u003ccode\u003e(\u0026lt;top:int\u0026gt;,\u0026lt;left:int\u0026gt;,\u0026lt;height:int\u0026gt;,\u0026lt;width:int\u0026gt;)\u003c/code\u003e.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_train_object_action_correspondence.csv\"\u003eEPIC_train_object_action_correspondence.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 5 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 12%\" /\u003e\n\u003ccol style=\"width: 5%\" /\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 72%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo the frames are part of.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eobject_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e56581\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame number of the object detection image from \u003ccode\u003eobject_detection_images\u003c/code\u003e.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eaction_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e56638\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame number of the corresponding image in the released frames for action recognition in \u003ccode\u003eframes_rgb_flow\u003c/code\u003e.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003etimestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:00:00.00\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eTimestamp in \u003ccode\u003eHH:mm:ss.SS\u003c/code\u003e corresponding to the frame.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePlease note we have included a python pickle file for ease of use. This includes a pandas dataframe with the same layout as above. This pickle file was created with pickle protocol 2 on pandas version 0.22.0.\u003c/p\u003e\n\u003ch3 id=\"epic_test_s1_object_action_correspondence.csv\"\u003eEPIC_test_s1_object_action_correspondence.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 5 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 12%\" /\u003e\n\u003ccol style=\"width: 5%\" /\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 72%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_11\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo containing the object s1 test frames.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eobject_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e33601\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame number of the object detection image from \u003ccode\u003eobject_detection_images\u003c/code\u003e.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eaction_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e33635\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame number of the corresponding image in the released frames for action recognition in \u003ccode\u003eframes_rgb_flow\u003c/code\u003e.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003etimestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:09:20.58\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eTimestamp in \u003ccode\u003eHH:mm:ss.SS\u003c/code\u003e corresponding to the frames.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePlease note we have included a python pickle file for ease of use. This includes a pandas dataframe with the same layout as above. This pickle file was created with pickle protocol 2 on pandas version 0.22.0.\u003c/p\u003e\n\u003ch3 id=\"epic_test_s2_object_action_correspondence.csv\"\u003eEPIC_test_s2_object_action_correspondence.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 5 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 12%\" /\u003e\n\u003ccol style=\"width: 5%\" /\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 72%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP09\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP09_05\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo containing the object s2 test frames.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eobject_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e15991\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame number of the object detection image from \u003ccode\u003eobject_detection_images\u003c/code\u003e.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eaction_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e16007\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame number of the corresponding image in the released frames for action recognition in \u003ccode\u003eframes_rgb_flow\u003c/code\u003e.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003etimestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:04:26.78\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eTimestamp in \u003ccode\u003eHH:mm:ss.SS\u003c/code\u003e corresponding to the frames.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePlease note we have included a python pickle file for ease of use. This includes a pandas dataframe with the same layout as above. This pickle file was created with pickle protocol 2 on pandas version 0.22.0.\u003c/p\u003e\n\u003ch3 id=\"epic_test_s1_object_video_list.csv\"\u003eEPIC_test_s1_object_video_list.csv\u003c/h3\u003e\n\u003cp\u003eCSV file listing the videos used to obtain the object s1 test frames. The frames can be obtained from \u003ca href=\"https://data.bris.ac.uk/data/dataset/3h91syskeag572hl6tvuovwv4d\"\u003eRDSF\u003c/a\u003e under \u003ccode\u003eobject_detection_images/test\u003c/code\u003e. Please test all frames from this folder for the videos listed in this csv.\u003c/p\u003e\n\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_11\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo containing the object s1 test frames.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_test_s2_object_video_list.csv\"\u003eEPIC_test_s2_object_video_list.csv\u003c/h3\u003e\n\u003cp\u003eCSV file listing the videos used to obtain the object s2 test frames. The frames can be obtained from \u003ca href=\"https://data.bris.ac.uk/data/dataset/3h91syskeag572hl6tvuovwv4d\"\u003eRDSF\u003c/a\u003e under \u003ccode\u003eobject_detection_images/test\u003c/code\u003e. Please test all frames from this folder for the videos listed in this csv.\u003c/p\u003e\n\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_11\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo containing the object s2 test frames.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_test_s1_timestamps.csv\"\u003eEPIC_test_s1_timestamps.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 7 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 3%\" /\u003e\n\u003ccol style=\"width: 9%\" /\u003e\n\u003ccol style=\"width: 75%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003euid\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e1924\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eUnique ID of the segment.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_11\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo the segment is in.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:00:00.000\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:00:01.890\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e1\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e93\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePlease note we have included a python pickle file for ease of use. This includes a pandas dataframe with the same layout as above. This pickle file was created with pickle protocol 2 on pandas version 0.22.0.\u003c/p\u003e\n\u003ch3 id=\"epic_test_s2_timestamps.csv\"\u003eEPIC_test_s2_timestamps.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 7 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 3%\" /\u003e\n\u003ccol style=\"width: 9%\" /\u003e\n\u003ccol style=\"width: 75%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003euid\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e15582\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eUnique ID of the segment.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eparticipant_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP09\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the participant.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP09_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo the segment is in.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:00:01.970\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_timestamp\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e00:00:03.090\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd time in \u003ccode\u003eHH:mm:ss.SSS\u003c/code\u003e of the action.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003estart_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e118\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eStart frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003estop_frame\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e185\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eEnd frame of the action (WARNING only for frames extracted as detailed in \u003ca href=\"#video-information\"\u003eVideo Information\u003c/a\u003e).\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePlease note we have included a python pickle file for ease of use. This includes a pandas dataframe with the same layout as above. This pickle file was created with pickle protocol 2 on pandas version 0.22.0.\u003c/p\u003e\n\u003ch3 id=\"epic_noun_classes.csv\"\u003eEPIC_noun_classes.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 3 columns:\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNote: a colon represents a compound noun with the more generic noun first. So pan:dust should be read as dust pan.\u003c/em\u003e\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 9%\" /\u003e\n\u003ccol style=\"width: 23%\" /\u003e\n\u003ccol style=\"width: 25%\" /\u003e\n\u003ccol style=\"width: 41%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e2\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the noun class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eclass_key\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003epan:dust\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eKey of the noun class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enouns\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003elist of string (1 or more)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e\u0026quot;['pan:dust', 'dustpan']\u0026quot;\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eAll nouns within the class (includes the key).\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_verb_classes.csv\"\u003eEPIC_verb_classes.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 3 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 9%\" /\u003e\n\u003ccol style=\"width: 22%\" /\u003e\n\u003ccol style=\"width: 29%\" /\u003e\n\u003ccol style=\"width: 38%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everb_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e3\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the verb class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eclass_key\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eclose\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eKey of the verb class.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everbs\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003elist of string (1 or more)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e\u0026quot;['close', 'close-off', 'shut']\u0026quot;\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eAll verbs within the class (includes the key).\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_descriptions.csv\"\u003eEPIC_descriptions.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing 4 columns:\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 10%\" /\u003e\n\u003ccol style=\"width: 5%\" /\u003e\n\u003ccol style=\"width: 39%\" /\u003e\n\u003ccol style=\"width: 44%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo_id\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eID of the video.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003edate\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e30/04/2017\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eDate on which the video was shot.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003etime\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e13:49:00\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eLocal recording time of the video.\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003edescription\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eprepared breakfast with soy milk and cereals\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eDescription of the activities contained in the video.\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_many_shot_verbs.csv\"\u003eEPIC_many_shot_verbs.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing the many shot verbs. A verb class is considered many shot if it appears more than 100 times in training. (NOTE: this file is derived from \u003ccode\u003eEPIC_train_action_labels.csv\u003c/code\u003e, checkout the \u003ca href=\"https://github.com/epic-kitchens/epic-many-shot-classes\"\u003eaccompanying notebook\u003c/a\u003e demonstrating how we compute these classes)\u003c/p\u003e\n\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everb_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e1\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the verb class\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003everb\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eput\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVerb corresponding to the verb class\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_many_shot_nouns.csv\"\u003eEPIC_many_shot_nouns.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing the many shot nouns. A noun class is considered many shot if it appears more than 100 times in training. (NOTE: this file is derived from \u003ccode\u003eEPIC_train_action_labels.csv\u003c/code\u003e, checkout the \u003ca href=\"https://github.com/epic-kitchens/epic-many-shot-classes\"\u003eaccompanying notebook\u003c/a\u003e demonstrating how we compute these classes)\u003c/p\u003e\n\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e3\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the noun class\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003etap\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNoun corresponding to the noun class\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_many_shot_actions.csv\"\u003eEPIC_many_shot_actions.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing the many shot actions. An action class (composed of a verb class and noun class) is considered many shot if BOTH the verb class and noun class are many shot AND the action class appears in training at least once. (NOTE: this file is derived from \u003ccode\u003eEPIC_train_action_labels.csv\u003c/code\u003e, checkout the \u003ca href=\"https://github.com/epic-kitchens/epic-many-shot-classes\"\u003eaccompanying notebook\u003c/a\u003e demonstrating how we compute these classes)\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 11%\" /\u003e\n\u003ccol style=\"width: 7%\" /\u003e\n\u003ccol style=\"width: 38%\" /\u003e\n\u003ccol style=\"width: 42%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eaction_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003e(int, int)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e(9, 84)\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric Pair of IDs, first the verb, then the noun\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003everb_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e9\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the verb class\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003everb\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003emove\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVerb corresponding to the verb class\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun_class\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eint\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e84\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNumeric ID of the noun class\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003enoun\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003estring\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003esausage\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eNoun corresponding to the noun class\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch3 id=\"epic_video_info.csv\"\u003eEPIC_video_info.csv\u003c/h3\u003e\n\u003cp\u003eCSV file containing information for each video\u003c/p\u003e\n\u003ctable\u003e\n\u003ccolgroup\u003e\n\u003ccol style=\"width: 11%\" /\u003e\n\u003ccol style=\"width: 7%\" /\u003e\n\u003ccol style=\"width: 38%\" /\u003e\n\u003ccol style=\"width: 42%\" /\u003e\n\u003c/colgroup\u003e\n\u003cthead\u003e\n\u003ctr class=\"header\"\u003e\n\u003cth\u003eColumn Name\u003c/th\u003e\n\u003cth\u003eType\u003c/th\u003e\n\u003cth\u003eExample\u003c/th\u003e\n\u003cth\u003eDescription\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003evideo\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003e(string)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003eP01_01\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eVideo ID\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003eresolution\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003e(string)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e1920x1080\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eResolution of the video, format is \u003ccode\u003eWIDTHxHEIGHT\u003c/code\u003e\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"odd\"\u003e\n\u003ctd\u003e\u003ccode\u003eduration\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003e(float)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e1652.152817\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eDuration of the video, in seconds\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr class=\"even\"\u003e\n\u003ctd\u003e\u003ccode\u003efps\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003e(float)\u003c/td\u003e\n\u003ctd\u003e\u003ccode\u003e59.9400599400599\u003c/code\u003e\u003c/td\u003e\n\u003ctd\u003eFrame rate of the video\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch2 id=\"file-downloads\"\u003eFile Downloads\u003c/h2\u003e\n\u003cp\u003eDue to the size of the dataset we provide scripts for downloading parts of the dataset:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"https://github.com/epic-kitchens/download-scripts/blob/master/download_videos.sh\"\u003evideos\u003c/a\u003e (750GB)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://github.com/epic-kitchens/download-scripts/blob/master/download_frames_rgb_flow.sh\"\u003eframes\u003c/a\u003e (320GB)\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"https://raw.githubusercontent.com/epic-kitchens/download-scripts/master/frames_rgb_flow/download_rgb.sh\"\u003ergb-frames\u003c/a\u003e (220GB)\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://raw.githubusercontent.com/epic-kitchens/download-scripts/master/frames_rgb_flow/download_flow.sh\"\u003eflow-frames\u003c/a\u003e (100GB)\u003c/li\u003e\n\u003c/ul\u003e\u003c/li\u003e\n\u003cli\u003e\u003ca href=\"https://github.com/epic-kitchens/download-scripts/blob/master/download_object_detection_images.sh\"\u003eobject annotation images\u003c/a\u003e (80GB)\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cem\u003eNote: These scripts will work for Linux and Mac. For Windows users a bash installation should work.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThese scripts replicate the folder structure of the dataset release, found \u003ca href=\"https://data.bris.ac.uk/data/dataset/3h91syskeag572hl6tvuovwv4d\"\u003ehere\u003c/a\u003e.\u003c/p\u003e\n\u003cp\u003eIf you wish to download part of the dataset instructions can be found \u003ca href=\"https://github.com/epic-kitchens/download-scripts\"\u003ehere\u003c/a\u003e.\u003c/p\u003e\n\u003ch2 id=\"video-information\"\u003eVideo Information\u003c/h2\u003e\n\u003cp\u003eVideos are recorded in 1080p at 59.94 FPS on a GoPro Hero 5 with linear field of view. There are a minority of videos which were shot at different resolutions, field of views, or FPS due to participant error or camera. These videos identified using \u003ccode\u003effprobe\u003c/code\u003e are:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e1280x720: \u003ccode\u003eP12_01\u003c/code\u003e, \u003ccode\u003eP12_02\u003c/code\u003e, \u003ccode\u003eP12_03\u003c/code\u003e, \u003ccode\u003eP12_04\u003c/code\u003e.\u003c/li\u003e\n\u003cli\u003e2560x1440: \u003ccode\u003eP12_05\u003c/code\u003e, \u003ccode\u003eP12_06\u003c/code\u003e\u003c/li\u003e\n\u003cli\u003e29.97 FPS: \u003ccode\u003eP09_07\u003c/code\u003e, \u003ccode\u003eP09_08\u003c/code\u003e, \u003ccode\u003eP10_01\u003c/code\u003e, \u003ccode\u003eP10_04\u003c/code\u003e, \u003ccode\u003eP11_01\u003c/code\u003e, \u003ccode\u003eP18_02\u003c/code\u003e, \u003ccode\u003eP18_03\u003c/code\u003e\u003c/li\u003e\n\u003cli\u003e48 FPS: \u003ccode\u003eP17_01\u003c/code\u003e, \u003ccode\u003eP17_02\u003c/code\u003e, \u003ccode\u003eP17_03\u003c/code\u003e, \u003ccode\u003eP17_04\u003c/code\u003e\u003c/li\u003e\n\u003cli\u003e90 FPS: \u003ccode\u003eP18_09\u003c/code\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThe GoPro Hero 5 was also set to drop the framerate in low light conditions to preserve exposure leading to variable FPS in some videos. If you wish to extract frames we suggest you resample at 60 FPS to mitigate issues with variable FPS, this can be achieved in a single step with FFmpeg:\u003c/p\u003e\n\u003cpre\u003e\u003ccode\u003effmpeg -i \u0026quot;P##_**.MP4\u0026quot; -vf \u0026quot;scale=-2:256\u0026quot; -q:v 4 -r 60 \u0026quot;P##_**/frame_%010d.jpg\u0026quot;\u003c/code\u003e\u003c/pre\u003e\n\u003cp\u003ewhere \u003ccode\u003e##\u003c/code\u003e is the Participant ID and \u003ccode\u003e**\u003c/code\u003e is the video ID.\u003c/p\u003e\n\u003cp\u003eOptical flow was extracted using a fork of \u003ca href=\"https://github.com/feichtenhofer/gpu_flow\"\u003e\u003ccode\u003egpu_flow\u003c/code\u003e\u003c/a\u003e made \u003ca href=\"https://github.com/dl-container-registry/furnari-flow\"\u003eavailable on github\u003c/a\u003e. We set the parameters: stride = 2, dilation = 3, bound = 25 and size = 256.\u003c/p\u003e\n\u003ch2 id=\"license\"\u003eLicense\u003c/h2\u003e\n\u003cp\u003eAll files in this dataset are copyright by us and published under the Creative Commons Attribution-NonCommerial 4.0 International License, found \u003ca href=\"https://creativecommons.org/licenses/by-nc/4.0/\"\u003ehere\u003c/a\u003e. This means that you must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. You may not use the material for commercial purposes.\u003c/p\u003e\n\u003ch2 id=\"disclaimer\"\u003eDisclaimer\u003c/h2\u003e\n\u003cp\u003eEPIC-KITCHENS-55 and EPIC-KITCHENS-100 were collected as a tool for research in computer vision, however, it is worth noting that the dataset may have unintended biases (including those of a societal, gender or racial nature).\u003c/p\u003e\n\u003ch2 id=\"changelog\"\u003eChangelog\u003c/h2\u003e\n\u003cp\u003eSee release history for changelog.\u003c/p\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fepic-kitchens%2Fepic-kitchens-55-annotations","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fepic-kitchens%2Fepic-kitchens-55-annotations","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fepic-kitchens%2Fepic-kitchens-55-annotations/lists"}